Research On Inverse Algorithm Of Electromagnetic Tomography

The physical fields of an aluminum electrolysis cell influence dramatically the efficiency of the electrical current and the working life of the cell. Hence, it is significant to simulate the muiti-physical fields inside the cell so as to optimize the fields and develop new reduction cells. Computer simulation techniques were applied using COMSOL multiphysics in this thesis, the electrical, magnetic and flow fields were computed for prebaked anode electrolysis cell where the electrical current is 300kA. An estimation method to evaluate the comprehensive performance of the aluminium reduction cells was presented, which can provide practical guide to the optimization of the cell structure and the development of new reduction cells.In the thesis, firstly, a mathematical model of flow field of molten aluminum was set up, k-ε dual equation model was used to describe the basic principles of the liquid aluminum flow. Besides, wall function method was adopted to treat the boundary condition. Secondly, forces imposed on liquid aluminum inside the 300kA-aluminum electrolysis cell were analyzed detailedly, which shows that the electromagnetic force is the main driving force of the circular flow of the liquid aluminum.Based on obtained data, the electromagnetic force fields were pre-computed using MATLAB programming, the results named as Fx.txt and Fy.txt were saved in the specified folder for further calling. And then, a finite element model for the aluminum flow was established by applying the finite element analysis software COMSOL Multiphysics. Adding a built-in function, the calculated electromagnetic force field was introduced as the solved volume force. Finally, computations were carried out and a variety of graphics were drawn by post-processing and visualization.Based on analyzing present situation of magnetic field optimization of the aluminum electrolysis cell, and taking account of the viewpoint of global optimization, the magnetic field of the aluminum electrolysis cell was optimized using genetic algorithm. In this paper, the principle of the basic genetic algorithm was introduced and its improvements were carried out, several optimization programs for the modified genetic algorithm were written using MATLAB language, whose high efficiency was validated by the optimization of two standard functions.